摘要
基于科技成果中试转化运行原理,搭建中试系统机理模型,利用改进GMDH构建相应数学模型。改进GMDH将含有回归项线性相关检测的快速递归算法(FRA)代替传统最小二乘(OLS)进行GMDH多层建模的部分多项式求解,使用初始变量添加法避免有效神经元过早丢失。以改进的GMDH对上海市中试转化统计数据进行建模及仿真实验,结果表明,所建模型比传统GMDH模型精度高,泛化能力强。
On the basis of analysis of the pilot process operation principle,this paper builds a pilot conversion mechanistic model and its mathematical model through using the improved batch data processing algorithm( GMDH); This paper replaces the traditional least squares algorithm with the fast recursive algorithm( FRA) containing a linear correlation detection to calculate the coefficients of the polynomial in the process of the multilayer modeling,and designs the initial variables to avoid active neurons losing early. The improved algorithm is applied to system modeling based on the Shanghai pilot conversion statistical data,the simulation results show that the accuracy and generalization capability of the built model are better than the traditional model.
引文
[1]陈肖夏,王晶.应用技术成果转化绩效系统评价[J].技术创新与管理,2009(3):313-316.CHEN Xiaoxia,WANG Jing.Performance evaluation system of achievements tranformation in applied technology[J].Technology and Innovation Management,2009(3):313-316.(in Chinese)
[2]鲍林,黄朗喜.科技成果转化的中试环节建设[J].研究与发展管理,2006,14(4):30-35.BAO Lin,HUANG Langxi.Technology transfer pilot link building[J].R and D Management,2006,14(4):30-35.(in Chinese)
[3]刘姝威,陈伟忠,王爽.提高我国科技成果转化率的三要素[J].中国软科学,2006(4):55-60.LIU Shuwei,CHEN Weizhong,WANG Shuang.The three main factors to raise the rate of China’s technology transfer[J].China Soft Science,2006(4):55-60.(in Chinese)
[4]王桂月,王树恩.基于多智能体的高校科技成果转化实施知识管理的组织形式研究[J].科技进步与对策,2010,27(3):145-148.WANG Guiyue,WANG Shuen.Research on knowledge management organization form of scientific and technological achievement transformation in universities based on multi-agent system[J].Science and Technology Progress and Policy,2010,27(3):145-148.(in Chinese)
[5]徐晨,邵云飞.基于DEA的科技成果转化绩效评价研究[J].电子科技,2010,23(7):58-61.XU Chen,SHAO Yunfei.Study of evaluation of transfer efficiency of scientific and technological achievements[J].Electronic Science and Technology,2010,23(7):58-61.(in Chinese)
[6]陈一君.科技成果转化系统分析及模式研究[J].科技情报与经济,2013,13(9):99-101.CHEN Yijun.System analysis and mode study on scientific and technical payoffs transformation[J].Sci/Tech Information Development and Economy,2013,13(9):99-101.(in Chinese)
[7]Lemke F,Muller J A.Self-organizing data mining[J].Systems Analysis Modeling Simulation,2003,43:231-240.
[8]张瑛.基于自组织数据挖掘的应用技术成果转化系统研究[D].南京:南京航空航天大学,2011.
[9]朱帮助,魏一鸣.基于GMDH-PSO-LSSVM的国际碳市场价格预测[J].系统工程理论与实践,2011,32(12):2264-2270.ZHU Bangzhu,WEI Yiming.Carbon price prediction based on integration of GMDH,particle swarm optimization and least squares support vector machines[J].Systems Engineering-Theory and Practice,2011,31(12):2264-2270.(in Chinese)
[10]张晓明,朱军.中试突破——成果转化系统模型设计与分析[J].科技成果纵横,2008(4):17-19.ZHANG Xiaoming,ZHU Jun.Pilot breakthrough-design and analysis of achievement transformation system[J].Technology Results Summarize,2008(4):17-19.(in Chinese)
[11]王晶,韩愈,詹莉.基于混合方法的“应用技术成果转化系统”的模型构建[C]//中国系统工程学会第17届年会论文集.镇江:江苏省系统工程学会,2012.
[12]Hastie T,Tibshirani R,Friedman J.The Elements of Statistical Learning:Data Mining,Inference,and Prediction[M].Heidelberg:Springer,2003.
[13]LI Kang,PENG Jianxun.A fast input selection algorithm for neural modeling of nonlinear dynamic systems[C]//Proceeding of the 2005 International Conference on Intelligent Computing.Heidelberg:Springer-Verlag,2005:1045-1054.
[14]LI K,PENG J,Irwin G.A fast nonlinear model identification method[J].IEEE Transactions on Automatic Control,2005,50(8):1211-1216.